The instant turn-around works *without* app-reloading. And I'm guilty of
liking it - without being a Ruby fan.
It's a feature of Eclipse and the JDK (>=1.4). Have a look at
Preferences/Java/Debug.

> -----Original Message-----
> From: Bryan Lewis [mailto:[EMAIL PROTECTED]
> Sent: Tuesday, January 17, 2006 6:30 PM
> To: Tapestry users
> Subject: Re: First page display slowdown in v 4.0
> 
> 
> I hear what you're saying!  I prefer the good old
> compiled-strong-type-checking Java approach... not a fan of perl,
> haven't tried ruby.  I don't even like annotations yet. :-)  But I saw
> some comments on the web that appeared to be  promising instant
> turn-around for simple edits and I wanted it if I could get 
> it.  Up till
> now I've been happy with instant turn-around for html/page 
> edits.  I can
> live with an auto-reload after a code change, now that I know that's
> what people are talking about.
> 
> Thanks for the clarification.
> 
> 
> Patrick Casey wrote:
> 
> >     That's the expected behavior. When you change a java 
> file, your IDE
> >recompiles the .class file. That, in turn, causes Tomcat (if 
> reloadable is
> >true) to throw away the current set of classes in memory and 
> do a reload.
> >For a simple project, this doesn't take long at all, but 
> when you have large
> >frameworks getting reloaded, be it Hibernate or Hivemind, 
> you're going to
> >take a performance hit.
> >
> >     A good rule of thumb is that if you're trying to 
> hot-patch live code
> >on a regular basis, take a step back and work a bit on your coding
> >practices. Especially if you come from a dynamic language 
> background like
> >perl or ruby it's tempting to just start hacking the code of 
> a running
> >program. That's not usually the best approach with java 
> though, despite the
> >fact that with a modern IDE and debugger you can work this way.
> >
> >     Generally speaking you're better off making a bunch of 
> changes at
> >once, doing a recompile, and testing, rather than trying to 
> hot fix them one
> >at a time.
> >
> >     --- Pat
> >
> >  
> >
> >>-----Original Message-----
> >>From: Bryan Lewis [mailto:[EMAIL PROTECTED]
> >>Sent: Monday, January 16, 2006 8:18 PM
> >>To: Tapestry users
> >>Subject: Re: First page display slowdown in v 4.0
> >>
> >>Yes, I tried that.  Whenever I changed a java file, the app 
> context got
> >>restarted which was rather slow.  Took 10 seconds for the app to be
> >>ready to use again.  Is that what's supposed to happen?  I 
> was hoping
> >>for something like a one-second reload of just the one file 
> that changed.
> >>
> >>
> >>Patrick Casey wrote:
> >>
> >>    
> >>
> >>>   Did you set reloadable="true" in your web.xml?
> >>>
> >>>   --- Pat
> >>>
> >>>
> >>>
> >>>      
> >>>
> >>>>-----Original Message-----
> >>>>From: Bryan Lewis [mailto:[EMAIL PROTECTED]
> >>>>Sent: Monday, January 16, 2006 5:29 PM
> >>>>To: Tapestry users
> >>>>Subject: Re: First page display slowdown in v 4.0
> >>>>
> >>>>I must be doing something wrong then.  I've tried three 
> app servers,
> >>>>currently trying JBoss/Tomcat.  Running inside Eclipse, 
> starting in
> >>>>debug mode.  I have Tapestry's caching disabled in the startup
> >>>>properties, and changes to the html or page files do take effect
> >>>>immediately.  But when I make a small change inside a 
> Java method... it
> >>>>does get built  automatically, and the class file does 
> get updated in
> >>>>the webapp tree.  But the change doesn't show up in the 
> browser until I
> >>>>restart the server or reload the app.
> >>>>
> >>>>
> >>>>Alexander Varakin wrote:
> >>>>
> >>>>
> >>>>
> >>>>        
> >>>>
> >>>>>Hot-code-replace feature is available in Eclipse if you run web
> >>>>>          
> >>>>>
> >>container
> >>    
> >>
> >>>>>          
> >>>>>
> >>>>in
> >>>>
> >>>>
> >>>>        
> >>>>
> >>>>>debug mode. The problem is that it takes twice longer to 
> display first
> >>>>>
> >>>>>
> >>>>>          
> >>>>>
> >>>>page
> >>>>
> >>>>
> >>>>        
> >>>>
> >>>>>in debug mode. Also Hot-code-replace works only if you 
> don't touch
> >>>>>
> >>>>>
> >>>>>          
> >>>>>
> >>>>function
> >>>>
> >>>>
> >>>>        
> >>>>
> >>>>>declarations.
> >>>>>
> >>>>>
> >>>>>
> >>>>>
> >>>>>
> >>>>>          
> >>>>>
> >>>>----------------------------------------------------------
> -----------
> >>>>To unsubscribe, e-mail: 
> [EMAIL PROTECTED]
> >>>>For additional commands, e-mail: 
> [EMAIL PROTECTED]
> >>>>
> >>>>
> >>>>        
> >>>>
> >>>
> >>>
> >>>-----------------------------------------------------------
> ----------
> >>>To unsubscribe, e-mail: 
> [EMAIL PROTECTED]
> >>>For additional commands, e-mail: 
> [EMAIL PROTECTED]
> >>>
> >>>
> >>>
> >>>      
> >>>
> >
> >
> >
> >
> >---------------------------------------------------------------------
> >To unsubscribe, e-mail: [EMAIL PROTECTED]
> >For additional commands, e-mail: 
> [EMAIL PROTECTED]
> >
> >  
> >
> 
> 

---------------------------------------------------------------------
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]

Reply via email to